Search in the Abstract Database

Abstracts Search 2019

OP27 High-dimensional mass cytometry reveals the immune cell landscape in inflammatory bowel disease

V. van Unen1, N. Li1, T. Abdelaal2,3, Y. Kooy-Winkelaar1, L. Ouboter*1,4, G. Beyrend1, T. Höllt3,6, L. Mearin7, A. Witte8, H. Escher9, B. Lelieveldt10,11, A. van der Meulen - de Jong4, F. Koning1

1Leiden University Medical Center, Department of Immunohematology and Blood Transfusion, Leiden, The Netherlands, 2Delft University of Technology, Delft Bioinformatics Lab, Delft, The Netherlands, 3Leiden University Medical Center, Leiden Computational Biology Center, Leiden, The Netherlands, 4Leiden University Medical Center, Department of Gastroenterology, Leiden, The Netherlands, 6Delft University of Technology, Computer Graphics and Visualization, Delft, The Netherlands, 7Leiden University Medical Center, Department of Paediatrics, Leiden, The Netherlands, 8Alrijne Hospital, Department of Gastroenterology, Leiderdorp, The Netherlands, 9Erasmus University Medical Center, Department of Paediatric Gastroenterology, Rotterdam, The Netherlands, 10Delft University of Technology, Pattern Recognition and Bioinformatics Group, Delft, The Netherlands, 11Leiden University Medical Center, Department of LKEB Radiology, Leiden, The Netherlands


Inflammatory bowel disease (IBD) is characterised by chronic inflammation of the intestine. Studies on individual immune lineages have shown alterations in the innate and adaptive intestinal immune system implicated in IBD. However, a comprehensive analysis of the cell composition in intestinal biopsies from IBD patients across all major immune lineages simultaneously was lacking.


In patients aged 10–40 years with a clinical suspicion of IBD, we took paired biopsies (N = 104) from ileum and colon (both inflamed and uninflamed mucosa if available) and blood samples in 23 IBD patients and in 15 controls with a normal colonoscopy. Single-cell suspensions were stained with a 36-antibody panel and analysed with mass cytometry. The generated dataset was analysed with Hierarchical t-SNE (HSNE) in the Cytosplore analysis and visualisation tool.


In total, we identified 309 distinct cell clusters from the collective intestinal dataset containing 3.4 million cells in a data-driven manner. Here, controls clustered separate from patients, ileum samples separate from colon samples, and affected segments separate from unaffected segments (Figure 1A). However, affected samples from the different subgroups of IBD (Crohn’s disease, ulcerative colitis, undeterminate colitis) were mostly intermixed, suggesting similarities in the immune profiles. Moreover, we observed a large interindividual variation in the immune cell composition, indicative of unique individual ‘immune fingerprints’ in the intestinal tract. In addition, 19 subsets were significantly different between affected-IBD samples and unaffected-IBD samples/controls. Finally, in a correlation analysis, several CD4+ T-cell clusters correlated with ILC and myeloid cell clusters and were up-regulated in IBD-affected segments (Figure 1B, top-left network), while in particular TCRgd cell clusters (Figure 1B, top-right network) and a group of ILC clusters (Figure 1B, bottom network) were up-regulated in unaffected samples of patients and controls.

Abstract OP027 – Figure 1. Integrated immune system analysis of immune cell infiltrates. (A) PCA-visualisation of samples clustered for 309 cell cluster frequencies. Coloured for clinical info. (B) Network representation of immune cell cluster correlations.


Our study provides evidence that a coordinated cellular network of both innate and adaptive immune cell types are implicated in IBD. Together with the evidence for the unique individual-specific composition of the intestinal immune system, this may aid in the development of more (cost-)effective and personalised patient care.